Constructing an Ecological Security Pattern Coupled with Climate Change and Ecosystem Service Valuation: A Case Study of Yunnan Province

Abstract

Ecosystem services provide the scientific foundation and optimization objectives for constructing ecological security patterns, and their spatial characteristics directly affect planning decisions such as ecological source identification and corridor layout. However, current methods for constructing ecological security patterns rely excessively on static spatial optimization of landscape structure and ecological processes, while overlooking the dynamic variations in ecosystem service values under climate change. Taking Yunnan Province as a case study, this paper calculates ecosystem service values, analyzes their spatiotemporal variations, and based on ecosystem service value hotspots, applies the MSPA model and circuit theory to identify ecological sources, corridors, pinch points, barrier areas, and improvement areas. On this basis, we construct and optimize the ecological security pattern of Yunnan Province and propose ecological protection strategies. The results show that: (1) From 2000 to 2030, ecosystem service values in Yunnan exhibit significant spatiotemporal heterogeneity. From 2000 to 2020, they first declined and then increased, with aquatic ecosystems contributing the most. Under future climate scenarios, ecosystem service values continue to increase, with the greatest growth under the SSP2-4.5 scenario. The spatial pattern is characterized by higher values in the central region and lower values in the eastern and western areas. (2) In 2020, 56 ecological sources were identified; under the SSP1-1.9 scenario, 61 were identified, while 57 were identified under both SSP2-4.5 and SSP5-8.5 scenarios. These sources are mainly distributed in northwestern Yunnan and the Nujiang and Lancang River basins, presenting a “more in the west, fewer in the east” pattern. (3) In 2020, 132 ecological corridors and 74 pinch points were identified. By 2030, under SSP1-1.9, there are 149 corridors and 84 pinch points; under SSP2-4.5, 135 corridors and 55 pinch points; and under SSP5-8.5, 134 corridors and 60 pinch points. (4) By integrating results across multiple scenarios, an ecological security pattern characterized as “three screens, two zones, six corridors, and multiple points” is constructed. Based on regional ecological background characteristics, differentiated strategies for ecological security protection of territorial space are proposed. This study provides a scientific reference for the synergistic optimization of ecosystem services and ecological security patterns under climate change.

1. Introduction

Ecological security refers to the fundamental environmental conditions that support human society by maintaining the balance, health, and sustainability of natural ecosystems. As a spatial planning tool for achieving ecological security, the Ecological Security Pattern (ESP) focuses on scientifically identifying key elements such as ecological nodes, corridors, and barriers, and constructing an ecological network with structural redundancy and functional resilience [1]. It is noteworthy that the ESP framework, although mainly developed and applied in Chinese research, shares fundamental principles with internationally recognized concepts such as Ecological Networks and Green Infrastructure, each aiming to safeguard core areas, create ecological corridors, and manage the matrix for holistic ecosystem conservation. This pattern not only effectively buffers disturbances caused by climate change and human activities but also coordinates the relationship between development and conservation through the delineation of ecological protection red lines, thereby providing essential ecological safeguards for regional sustainable development [2,3,4].

Land constitutes the essential material base and spatial platform for human survival and socio-economic development. Alterations in land use patterns result from the interplay between natural environmental changes and human activities, significantly influencing the structural characteristics and functional services of ecosystems [5,6]. In recent years, intensified human activities have led to global ecological degradation and resource depletion, seriously threatening sustainable development [7,8]. Against this background, the theory of Ecosystem Services (ES) emerged, emphasizing the comprehensive value of material supply, environmental regulation, and cultural support provided by natural systems to mankind, which is the key to coordinating ecological protection and social development [9]. However, rapid urbanization and excessive land development have caused a sharp decline in ecosystem functions [10,11], and the sharp decline in biodiversity not only destroys ecological balance but also directly threatens human welfare and is contrary to the green development goals [12].

Integrating ecosystem service values with land use has become a major focus of current research. However, accelerated urbanization has intensified global warming, and climate change alters vegetation growth processes and ecosystem functions, increasing the uncertainty and complexity of future land use and ecological security patterns [13,14]. To enhance the precision of global change simulations, the World Climate Research Programme (WCRP) initiated the Coupled Model Intercomparison Project (CMIP), which progressively developed new climate–socio-economic scenario frameworks, namely the Representative Concentration Pathways (RCPs), the Shared Socio-economic Pathways (SSPs), and their integrated SSP–RCP combinations [15,16]. In its latest phase, CMIP6, the Intergovernmental Panel on Climate Change (IPCC) merges SSPs with RCPs to generate a set of multi-scenario projections for future global climate change [17]. A growing body of research has applied SSP–RCP data to simulate alternative development trajectories [18,19]. Therefore, incorporating ecosystem service valuation into such climate–socio-economic scenarios [20] provides both a theoretical basis and practical guidance for assessing ecosystem services and ecological security under future conditions.

With the increasing trend of interdisciplinary integration, the study of ecological security patterns has gradually formed a comprehensive framework covering landscape pattern dynamics, ecological process simulation, and disturbance factor analysis [21,22]. The methodology has shifted from early GIS-based spatial overlay analysis to complex simulations with multiple model coupling, such as circuit theory, weighted average (OWA), minimum cumulative resistance model (MCR), etc. [1,23,24]. Among them, Yu et al. [25] innovatively proposed the “point line surface” spatial configuration theory of ecological security pattern in their research. By quantifying the interactions between ecological source areas and landscape elements, this theory has become the mainstream paradigm in the field and is widely applied to regional ecological security pattern research. Qin et al. [26] constructed the ecological security pattern of Kunming city, innovatively integrating the InVEST model with assessments of landscape connectivity and circuit theory, and identified critical zones for ecological restoration of national territorial space; Cai et al. [27] predicted the value of ecosystem services based on climate change and constructed an ecological security pattern, providing valuable insights and references for future decisions on ecosystem protection and land resource management. However, there are still some shortcomings: Firstly, referring to the research results of Xie et al. [28,29]. to analyze the value of ecosystem services, the study scope is generally on plains or watersheds, and there is relatively little study on the plateau mountainous areas; Secondly, existing analyses are mostly based on the current land use situation, and there is insufficient characterization of the dynamic mechanism of how climate change factors affect service value; Finally, current research on ESPs is mostly limited to local scales such as cities and counties, and is based on static conditions, lacking study on the distribution of ESPs in plateau mountainous areas provinces or future development scenarios.

Considering the study area’s future development plan, this study selects three SSP-RCP scenarios: SSP1-1.9, SSP2-4.5, and SSP5-8.5, representing sustainable development, intermediate, and fossil fuel–dominated pathways, respectively, for comprehensive analysis. This approach aims to provide new insights into the construction of ecological security patterns and ecological conservation in the study area. Numerous studies on multi-scenario land use simulation can generally be categorized into quantity optimization simulations and spatial distribution simulations. Common methods for designing land demand include the SD model, linear programming, and the Markov model [30,31]. While widely used, traditional approaches like the Markov model often involve subjectivity [27]. The SD model can integrate system complexity, nonlinear effects, and temporal relationships, and demonstrates significant advantages in simulating quantitative and structural changes of urban land use, scenario analysis, and decision support [32]. To simulate spatial distribution patterns, this study adopts the advanced Patch-generating Land Use Simulation (PLUS) model. This model incorporates roulette competition and adaptive inertia mechanisms, and uses the Random Forest (RF) algorithm to evaluate development potential across land types. This provides a comprehensive understanding of land use dynamics while improving accuracy and robustness. By coupling the SD and PLUS models, this study integrates land use quantity and spatial distribution for combined simulation, enabling effective prediction of land use changes and providing strong support for future research on the spatiotemporal patterns of land use.

Therefore, this study integrates multiple methods—including multi-scenario land use simulation (SD-PLUS model), ecosystem service value (ESV) assessment, circuit theory, and the morphological spatial pattern analysis (MSPA). This integration allows for the quantitative measurement of ESV and the systematic identification of ecological sources, corridors, and key nodes. It balances the scientific rigor of functional assessment with the systematicity of spatial simulation. Furthermore, by incorporating future scenario analysis, it reveals the ESV and ESPs under three climate scenarios for 2030, offering theoretical and methodological support for regional ecological conservation and spatial planning. The study takes Yunnan Province, a typical plateau and mountainous region, as its case study to demonstrate this applied research framework.

Yunnan Province constitutes a critical ecological barrier for environmental security and sustainable development in southwestern China and Southeast Asia. However, rapid urbanization and land use restructuring have exacerbated climate-related pressures. The pronounced topographic heterogeneity of this plateau-mountain region further induces significant climatic differentiation, intensifying ecological degradation through severe soil erosion, rocky desertification, and biodiversity loss. These challenges substantially impede regional sustainable development. To address these issues, this study evaluates ecosystem service values and constructs ecological security patterns under historical and future climate scenarios. The findings inform targeted conservation strategies, supporting the optimization of Yunnan’s territorial development system, enhanced ecological space management, and improved coordination between land use and ecosystem services.

2. Materials and Methods

2.1. Overview of the Study Area

Yunnan Province, abbreviated as “Dian”, is located in the plateau mountainous region of southwestern China, between latitude 21°8′ N and 29°15′ N and longitude 97°31′ E and 106°11′ E, with a total area of 384,400 km², and has the longest land border of any Chinese province. Elevations are higher in the northwest and progressively lower toward the southeast, decreasing step by step from north to south, with an elevation difference exceeding 6000 m. The climate is complex and varied, characterized by monsoon climate, low-latitude climate, and plateau climate, exhibiting obvious spatial gradients in precipitation that gradually decrease annually from south to north. There are 1002 rivers with a watershed area of more than 100 km², belonging to the six major water systems of Jinsha River, Nanpan River, Yuan River, Lancang River, Nujiang River, and Ayeyarwady River, of which the Lancang River, the Nujiang River, and the Yuan River are of great geo-ecological significance as transboundary rivers. This province possesses abundant biodiversity and is often referred to as the “Kingdom of Flora and Fauna.” Due to the interaction of climate, biology, geology, and topography, a variety of soil types have been formed, and the vertical distribution of soil is obvious, earning the name of “Red Soil Plateau”. This unique natural environment provides an important foundation for biodiversity conservation, water retention, soil retention, and the maintenance of other crucial ecosystem services.

With its location advantages, ecological functions, and cultural diversity, Yunnan has become a key region for global ecological protection and sustainable development, and has significant advantages in the fields of economy, trade, and tourism. However, activities such as hydroelectric development, mineral extraction, and tourism expansion have exacerbated forest fragmentation and habitat loss, which have triggered ecological degradation and a biodiversity crisis. How to balance ecological protection and regional development has become a key issue in Yunnan’s ecological civilization construction (Figure 1).

2.2. Data Sources and Processing

The data used in this study primarily include land use data, geographic data, climate data, and socio-economic data. The specific data sources are shown in

Table 1. Data Sources.

Data Name	Data Description	Spatial Resolution	Data Sources
Land use data	Land use data	30 m	Resource and environmental science data platform (https://www.resdc.cn/) accessed on 16 October 2024.
Climate data	Temperature, precipitation, and potential evapotranspiration	1 km
Socio-economic data	GDP, POP, grain output, grain sown area	/	“Compilation of Agricultural Cost and Benefit in China”, Statistical Yearbook of Yunnan Province” accessed on 18 October 2024.
	Grain price	yuan/hm2	“Grain Statistical Yearbook of Yunnan Province” (accessed on 18 October 2024).
Geographic data	Road network data	/	OpenStreetMap (www.openstreetmap.org) accessed on 11 October 2024.
	River data	/
	DEM	30 m	Geospatial data cloud (www.gscloud.cn) accessed on 13 October 2024.
	Slope and aspect	/	Obtained through DEM processing
	GDP and POP in 2030	1 km	Science data bank (https://cstr.cn/31253.11.sciencedb.01683) [33,34,35] (accessed on 7 October 2024).
SSP-RCP dataset	Temperature, precipitation, and potential evapotranspiration in 2030	1 km	National Earth System Science Data Center (http://www.geodata.cn/main/) accessed on 18 October 2024.

. Additionally, standardized preprocessing was carried out for these data. First, land use datasets for Yunnan Province in 2000, 2010, and 2020 were reclassified following the first-level classification system of the Chinese Academy of Sciences and subsequently extracted using the study area mask, resulting in six unified categories: cultivated land, forest, grassland, water bodies, construction land, and unused land. Meanwhile, based on the vector data of Yunnan Province, combined with the CMIP6 climate data, the driving factor data were collated: the slope and aspect factors of the study area were extracted using the DEM data; POP and GDP raster data were processed by projection transformation, cropping, and resampling; the spatial distribution data of temperature and precipitation were obtained through mask extraction; and the Euclidean distance analysis was applied to obtain the distance raster from the roads of all levels and the water system. In addition, yield, and price data of major grain crops (rice, wheat, and corn) in Yunnan Province were compiled to calculate average yields and unit prices. Finally, all data were standardized to a 300 m × 300 m spatial resolution using a unified projection (WGS_1984_UTM_ZONE_48N) and processed through masking, extraction, and resampling procedures.

2.3. Research Method

2.3.1. Research Framework

The overall research framework of this study is illustrated in Figure 2. First, using land use, geographic, socio-economic, and climate data from 2000, 2010, and 2020, integrated with the SSP-RCP scenario dataset, we project future land use changes under various climate scenarios. Second, based on both historical and projected land use data, we apply a modified equivalent factor method to assess ecosystem service values and analyze their spatial and temporal evolution under historical and projected scenarios. Finally, considering the characteristics of ecosystem service value hotspots and cold spots, we employ the Morphological Spatial Pattern Analysis (MSPA) model and circuit theory to identify ecological corridors, pinch points, and barrier points under historical and future scenarios. Based on this, we developed and refined the ecological security pattern for Yunnan Province, which led to the development of targeted ecological conservation strategies. The overall aim is to guide Yunnan’s territorial spatial planning from “incremental expansion” toward “stock optimization,” enhance ecosystem resilience, support the modernization of spatial governance, and promote equitable distribution of ecological benefits.

2.3.2. Land Use Simulation Method

1.

System dynamics model (SD model)

The SD model can capture the complex nonlinear relationships and feedback mechanisms between socio-economic driving factors, making it suitable for long-term quantitative land use quantity demand simulation in various scenarios. The Vensim(Version 10.2.0)-based SD model developed in this study consists of four interacting subsystems: economic, demographic, climatic, and land use (Figure 3). The economic subsystem simulates the impacts of industrial investment, real estate development and economic restructuring on land use types; the demographic subsystem reflects the pressure of scale expansion and consumption growth on construction and cultivated land; the climate subsystem quantifies the regulating effect of temperature and precipitation changes on ecological land use; and the land use subsystem integrates multiple drivers from diverse sources to simulate land use evolution. The model utilizes land use, socio-economic, and climate datasets covering the period from 2000 to 2020, and the variable relationships are verified by SPSS 25 to establish a system of state equations, and the simulation accuracy is verified with 2020 data. A one-year time step is set to simulate the change of land use quantity demand under different development scenarios from 2010 to 2030, focusing on predicting the land use pattern in 2030.

2.

Path-generating Land Use simulation (PLUS model)

The PLUS model is an advanced land use change simulation framework based on raster data that improves simulation accuracy by integrating the advantages of multiple algorithms [36]. Its core innovations include: (1) an adaptive inertia competition mechanism and a roulette wheel selection mechanism to optimize the competition among land use types; (2) the application of a random forest (RF) algorithm to precisely identify driving factors and generate development probability surfaces; (3) a land expansion analysis strategy (LEAS) that constructs a transformation rule mining framework; (4) a multi-type random seed mechanism (CARS) that improves upon conventional cellular automata (CA) models. By adopting a hybrid “top-down and bottom-up” simulation strategy, the PLUS model influences the generation of land use patches through adaptive coefficients constrained by development probability, thereby enabling the simulation of future land use distribution. These features allow the PLUS model to more effectively simulate the spontaneous growth of multiple land use types and accurately capture landscape-level changes, particularly in complex plateau mountain terrains.

3.

Future Scenario Setting

CMIP6 constructs an integrated scenario framework that combines socio-economic development and climate response by SSPs and RCPs. In this paper, three typical scenarios—SSP1-1.9 (sustainable development), SSP2-4.5 (intermediate pathway), and SSP5-8.5 (fossil fuel dominated)—are selected to represent, respectively, a green and low-carbon transition, the continuation of the historical trends, and a high-carbon development pathway. These scenarios quantitatively assess the future differential response of Yunnan Province in terms of ecosystem services, resource utilization, and climate adaptive capacity from the perspective of different socio-economic development modes and climate policy intensities and provide a scientific basis for decision-making on regional sustainable development.

2.3.3. Ecosystem Service Value (ESV) Assessment Methods

• Correction of the ESV equivalent factor

This paper is based on Xie et al. [29] “China’s Ecosystem Service Value Equivalent Scale per Unit Area” to estimate the value of ecosystem services in Yunnan Province. However, this method is based on the average level of ecosystems in China, which cannot fully reflect the differences and specialties of ecosystems. Therefore, we revised the equivalence factor value table according to the specific conditions of Yunnan Province in three aspects: land use types and grain yield, farmland and precipitation, and socio-economics, respectively.

• Step 1: Land Use Type and Grain Yield

  (1)

  Land use type correction

  The original equivalence table [29] contains 14 secondary classifications. In this study, the land use types in Yunnan Province were grouped into six major classes, namely cultivated land, forest, grassland, water bodies, construction land, and unused land. These categories were classified and adjusted based on the natural conditions and spatial distribution of land use in the province. Specifically, cultivated land represents the average of dry land and paddy fields; forest represents the average of all forest types; grassland represents the average of all grassland types; the average of water systems and glacier snow; and unused land represents the average of wetlands, deserts, and bare land. Construction land, meanwhile, is evaluated with reference to the study by Zhang et al. [37].

  (2)

  Grain yield correction

  Using the ecosystem correction method proposed by Xie et al. [28], we selected 1/7 of the economic value per unit area of three main crops—rice, wheat, and corn—in the study area to calculate ecosystem services per unit area.

• Step 2: Cultivated Land and Precipitation

The average grain yield is related to biomass, while rainfall is related to water resource supply and hydrological regulation services. We can revise the value equivalent table by studying the rainfall and grain yield in the study area.

• Step 3: Socio-economic

Referring to the existing research results [21], the ecosystem service value equivalent factor is dynamically corrected based on regional socio-economic characteristics. The correction method comprehensively considers the two dimensions of willingness to pay and ability to pay: willingness to pay is calculated through the relative value of GDP and population ratio between Yunnan Province and the whole country; ability to pay is quantified by constructing the social development stage coefficient in combination with urban and rural Engel coefficients and demographic structure; and finally, the regionalization of ecosystem service value is realized through the socio-economic correction coefficients. This method effectively reflects the differentiated impacts of human activities on ecosystem services at different stages of development.

2.

ESV Assessment

Based on the revised ESV equivalent of Yunnan Province, the ecosystem service value was calculated according to the formula.

$$ES{V}_i=\sum S_i\times E_i$$

(1)

ESV refers to the total value of the i-th ecosystem service in Yunnan Province; S_i is the area of the i-th land use type in Yunnan Province; and E_i is the amount of ecosystem service value per unit area.

2.3.4. Spatial Heterogeneity Analysis Methods

1.

Global Autocorrelation

The spatial autocorrelation approach assesses the spatial dependence of ecosystem service values (ESVs) using the global Moran’s I index, which ranges from −1 to 1: a positive value indicates agglomeration, a negative value indicates dispersion, and a zero value is a random distribution; Z-value and p-value are used to test the significance, and the larger the Z-value indicates the more significant the positive spatial correlation. The formula is as follows:

$$I={\displaystyle \frac{{\displaystyle \sum _{i=1}^n}{\displaystyle \sum _{j=1}^n}{\omega }_{ij}(x_i-\overline{x})(x_j-\overline{x})}{S^2{\displaystyle \sum _i}{\displaystyle \sum _j}{\omega }_{ij}}}$$

(2)

$$S^2={\displaystyle \frac{1}{n}}{\displaystyle \sum _{i=1}^n}{(x_i-\overline{x})}^2$$

(3)

where n is the total number of grid cells; x_i, x_j are the measured values of grid cells i, j; $(x_i-\overline{x})$, $(x_j-\overline{x})$ are the deviations of the measured values from the mean on the i, j grid cells; ${\omega }_{ij}$ is the standardized spatial weight matrix; S² is the variance.

2.

Cold and Hot Spot Analysis

The global Moran index is used to assess the spatial autocorrelation of ESVs, while the cold and hot spot analysis reveals their spatial agglomeration characteristics. In this paper, we apply the Getis-Ord statistic to reflect the distribution map of spatial clustering areas with significant high and low values of ecosystem services in Yunnan Province. The formula is given as follows:

$$G_i^{\ast }={\displaystyle \frac{{\displaystyle \sum _{j=1}^n}{\omega }_{ij}{x}_j-\overline{X}{\displaystyle \sum _{j=1}^n}{\omega }_{ij}}{\sqrt{\frac{{\displaystyle \sum _{j=1}^n}{x}_j^2}{n}-{(\overline{X})}^2}\sqrt{\frac{n{\displaystyle \sum _{j=1}^n}{\omega }_{ij}^2-{\left({\displaystyle \sum _{j=1}^n}{\omega }_{ij}\right)}^2}{n-1}}}}$$

(4)

where ${\omega }_{ij}$ denotes the direct spatial weight matrix of elements i and j, x_j is the attribute value of element j, $\overline{X}$ represents the mean of the attribute, while n denotes the total number of elements.

2.3.5. Ecological Security Pattern (ESP) Construction Method

1.

Ecological Source Area Identification Method

This paper adopts ESV assessment and MSPA [38,39,40] to identify ecological source areas in Yunnan Province. Firstly, based on the ESV hotspot analysis of patches (at the 90% confidence level), patches larger than 50 km² and overlapping with the nature reserves were selected as candidate ecological source areas. Secondly, Conefor software (Version 2.6) was used to calculate the landscape connectivity index (dPC value) [41], and the patches with the top 10% dPC value were classified as primary ecological source areas (key nodes), and the patches with 10–30% were classified as secondary ecological source areas (important corridors). This method provides a scientific basis for the construction of a regional ecological security pattern by integrating ecosystem service function and landscape connectivity [42]. The formula is as follows:

$$PC={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle }\sum _{j=1}^n}{m}_i{m}_j{p}_{ij}^{\ast }}{M_L^2}}$$

(5)

$$dpc={\displaystyle \frac{PC-P{C}_{remove}}{PC}}\times 100\%$$

(6)

where n denotes the total number of patches within the study area, m_i and m_j are the maximum product of all path probabilities between patches i and j, respectively; $M_L^2$ is the total area of the landscape in the study area; PC is the possible connectivity index of a patch; $dpc$ indicates the importance of plaque, and the greater the value, the higher the importance of plaque; PC_remove is the possible connectivity index after removing a patch.

2.

Resistance Surface Construction Method

In this paper, based on the environmental characteristics of Yunnan Province, eight key factors, namely, vegetation cover (NDVI), soil type (TRLX), slope (PD), distance from roads (DL), gross domestic product (GDP), population density (POP), temperature (QW) and precipitation (JS), were selected to construct the ecological resistance surface index system (

Table 2. Resistance factor indicator system for Yunnan Province (2020 and 2030).

Resistance Factor		Year	Grade					Weight
			10	30	50	70	90
Natural factor	NDVI	2020	>0.83	0.75~0.83	0.61~0.83	0.4~0.61	<0.4	0.2699
		2030	>0.83	0.75~0.83	0.61~0.83	0.4~0.61	<0.4
	TRLX	2020	201~237	1~35	181~201	35~139	139~181	0.2066
		2030	201~237	1~35	181~201	35~139	139~181
	PD	2020	<10	10~13	13~20	20~29	>29	0.0971
		2030	<10	10~13	13~20	20~29	>29
	QW	2020	<7	7~12	12~15.5	15.5~18.5	>18.5	0.0362
		2030	<6.5	6.5~12	12~16	16~19	>19
	JS	2020	>1900	1400~1900	1000~1400	850~1000	<850	0.0569
		2030	>1600	1300~1600	1100~1300	850~1100	<850
Social factor	DL	2020	>11,000	7300~11,000	4200~7300	1200~4200	<1200	0.0795
		2030	>11,000	7300~11,000	4200~7300	1200~4200	<1200
	GDP	2020	<760	760~3600	3600~13,000	13,000~33,000	>33,000	0.0455
		2030	<12,000	12,000~23,000	23,000~37,000	37,000~60,000	>60,000
	POP	2020	<0.02	0.02~0.1	0.1~0.28	0.28~0.45	>0.45	0.2083
		2030	<0.59	0.59~2.00	2.00~5.80	5.80~16.00	>16.00

). The factors were classified into natural and social criteria by the Analytic Hierarchy Process (AHP) (realized by Yaahp software (Version V7.5)), and the weight coefficients were determined. The resistance values were graded from 10 to 90 (10 is the optimal access condition), and the 2020 baseline data and 2030 SSP-RCP scenario data (GDP, population, etc.) were standardized using the natural breakpoint method. Finally, multi-period ecological resistance surfaces were generated through spatial overlay analysis (Figure 4).

3.

Ecological Corridors, Pinch Points, and Barrier Points Identification Method

Circuit theory can simulate the accessibility and fluidity of ecological processes throughout a region, in order to identify key nodes such as bottlenecks and obstacles, providing more detailed and actionable guidance for ecological restoration planning. In this study, based on circuit theory, the Linkage Mapper software and Circuitscape program (Version 4.0.5) were comprehensively applied to identify ecological corridors, pinch points, and obstacle zones [43,44]. Specifically: (1) Ecological corridors are identified using least-cost path analysis, with a cumulative resistance threshold set at 50,000 and high-resistance interference areas excluded; (2) Barrier Mapper is used to identify ecological obstacle points and optimize the setting of the search radius; and (3) the All-to-One model of Pinchpoint Mapper is applied to identify ecological pinch points using the current density method.

3. Results

3.1. Land Use Simulation Results

3.1.1. Accuracy Inspection

This study employs the SD-PLUS model, using land use data and driving factors from 2000 and 2010 for model training, and data from 2020 for model calibration and validation.

Table 3. Accuracy Validation of the SD-PLUS Model.

Land Use Type	Simulated Data (hm2)	Truthful Data (hm2)	Accuracy	Overall Accuracy
Cultivated land	6,710,640	6,770,736	−0.89%	93.00%
Forest	22,047,100	22,042,791	0.02%
Grassland	8,661,610	8,615,421	0.54%
Water area	372,580	380,061	−1.97%
Construction land	481,293	476,379	1.03%
Unused land	168,157	155,970	7.81%

summarizes the performance of the models. For each land use category, the SD model produced simulation errors below 5%, with an overall accuracy of 93%. Spatial accuracy assessment indicated that the PLUS model achieved a Kappa coefficient of 0.88. In addition, the simulation results of the land use in 2020 are basically the same as the actual land use distribution, and the simulation results are more compact, which achieves a high simulation performance. It shows that the model can accurately reflect the quantitative dynamics and spatial pattern characteristics of the land use system (Figure 5).

3.1.2. Spatial Distribution of Land Use

Land use in Yunnan Province from 2000 to 2030 shows vertical distribution characteristics (Figure 6). Forest land, which constitutes over 50% of the total area, is primarily distributed in southern and northwestern Yunnan; grassland is mainly distributed in Northwest Yunnan and some secondary urban agglomerations; cultivated land is concentrated in Central Yunnan and the tropical areas in the south; waters are distributed in Central Yunnan and Northwest Yunnan in the pattern of the “six parallel rivers”; construction land is focused on the urban agglomerations in Central Yunnan, and land use in the future under the three different climate scenarios will show an expansion pattern with Kunming, Qujing and Yuxi as the core circles respectively. The unique geographic and climatic conditions have shaped Yunnan’s ecological status as the “Water Tower of Asia,” providing a critical foundation for the region’s sustainable development.

3.2. Analysis of the Spatio-Temporal Variation Characteristics of ESVs

3.2.1. Temporal Variation Characteristics of ESVs

Based on the area of land use in Yunnan Province, the ESV for each land use type and the total ESV were calculated (

Table 4. Total ESVs in Yunnan Province from 2000 to 2030 (Million Yuan).

ES Classification	2000	2010	2020	2030
				SSP1-1.9	SSP2-4.5	SSP5-8.5
supply services	0.5357	0.5742	0.8352	0.9682	0.9883	0.9788
Regulating services	67.2451	67.1725	68.5754	68.8325	68.9182	68.9526
supporting service	1.1376	1.1394	1.1374	1.138	1.138	1.1381
cultural services	0.2267	0.2272	0.2279	0.2285	0.2286	0.2286
ecosystem services value	69.1452	69.1133	70.7759	71.1671	71.2731	71.298

and

Table 5. Total ESVs by Land Use Type in Yunnan Province from 2000 to 2030 (Million Yuan).

Land Use Type	2000	2010	2020	2030
				SSP1-1.9	SSP2-4.5	SSP5-8.5
cultivated land	0.9115	0.9079	0.8941	0.8801	0.8791	0.8802
forest	49.0943	49.4927	49.3132	49.2873	49.2801	49.2862
grassland	12.4479	12.1587	12.0549	12.1	12.096	12.1003
water area	7.2064	7.3495	9.7433	10.4324	10.6119	10.5605
construction land	−0.5304	−0.8069	−1.2408	−1.5399	−1.6001	−1.5361
unused land	0.0154	0.0114	0.0113	0.0072	0.006	0.0069
summation	69.1452	69.1133	70.7759	71.1671	71.2731	71.298

). The ESV showed a downward and then upward trend from 2000–2030, from 6914.52 million yuan to 7429.80 million yuan. The ESV was ranked as regulating services > supporting services > provisioning services > cultural services, with forest and water area as the main contributing land types.

Between 2000 and 2020, the ecosystem service value (ESV) of Yunnan Province increased by 163.07 million yuan, showing a decrease at first, followed by a later increase. Water (+253.68 million yuan) and forest (+21.89 million yuan) were the primary contributors to ESV growth, whereas the growth of construction land caused declines in the ESV of cultivated and grassland areas. During this period, provisioning, regulating, and cultural services increased by 29.94, 133.03, and 0.12 million yuan, respectively. After 2020, under the three future climate scenarios (SSP1-1.9, SSP2-4.5, and SSP5-8.5), ESVs are projected to increase by 39.12, 49.72, and 52.21 million yuan, respectively, driven by ecological protection policies. The SSP5-8.5 scenario exhibits the largest increase (52.21 million yuan), indicating that scientific and technological advances have partially offset the ecological damage caused by economic development. All scenarios show the most significant ESV growth in water, following the order SSP2-4.5 > SSP5-8.5 > SSP1-1.9.

3.2.2. Spatial Variation Characteristics of ESVs

As shown in Figure 7, the spatial distribution of ecosystem service value (ESV) in Yunnan Province from 2000 to 2030 remains generally consistent, exhibiting a pronounced clustering pattern. High-value areas (high or relatively high security zones) are mainly located in the northwestern and southwestern regions of Yunnan, along with adjacent eastern and western areas, which are rich in forest and grassland resources.

Between 2000 and 2020, due to rapid urbanization, the low security zone of the urban agglomeration in central Yunnan with Kunming as the core expanded significantly, while the medium security zone shifted toward western and northeastern Yunnan, and the high security zone remained stable in the northwestern and southwestern regions of Yunnan, which are rich in forest resources; The prediction of three climate scenarios in 2030 shows that SSP1-1.9 scenario maintains the widest high security zone range due to strict environmental protection policies, the SSP2-4.5 scenario, due to moderate environmental protection efforts, results in the largest expansion of the low security zone (including new cities such as Dali and Lijiang), while the SSP5-8.5 scenario, despite the most pronounced urban expansion (including new cities such as Zhaotong), benefits from technological progress facilitating the recovery and expansion of high security zones in the Lancang River and Yuan River basins, fully illustrating the dynamic game relationship between urbanization processes and ecological protection policies.

3.3. Spatial Heterogeneity Analysis of ESVs

3.3.1. Global Spatial Autocorrelation

As shown in

Table 6. Moran’s I Values of Primary ESVs in Yunnan Province from 2000 to 2030.

Services Type	2000	2010	2020	2030
				SSP1-1.9	SSP2-4.5	SSP5-8.5
total ESV	0.6388	0.6386	0.6391	0.6398	0.6399	0.6397
provisioning services	0.2907	0.2894	0.2872	0.2813	0.2834	0.2823
regulating services	0.3387	0.3361	0.3390	0.3402	0.3399	0.3398
supporting services	0.3378	0.3353	0.3376	0.3381	0.3380	0.3378
cultural services	0.3362	0.3338	0.3360	0.3361	0.3362	0.3360

, the ESVs in Yunnan Province from 2000 to 2030 exhibit significant spatial clustering characteristics (Moran’s I > 0.6386). The spatial autocorrelation of total ESV remains stable, with Moran’s I values fluctuating between 0.6386 and 0.6399. Among the various service categories, regulating services (Moran’s I > 0.3300) exhibit the highest and most stable spatial clustering, followed by supporting services and cultural services. In contrast, provisioning services (Moran’s I > 0.2800) show the lowest spatial clustering and declining trend. It is worth noting that, except for provisioning services, Moran’s I value of the other three service categories all exhibit a pattern of first decreasing and then increasing, maintaining a higher level of spatial clustering under all three climate scenarios by 2030, with the most significant clustering effect observed under the SSP2-4.5 scenario (Moran’s I = 0.6399). This spatial pattern evolution clearly reflects how human activities and natural conditions jointly shape the spatial distribution of ecosystem services.

3.3.2. Cold and Hot Spots Analysis

Figure 8 illustrates the spatial distribution characteristics of ecosystem service values (ESVs) in Yunnan Province from 2000 to 2030. Hotspots, which account for 30% of the total area, are primarily located in the Hengduan Mountains of Northwest Yunnan (Gaoligong, Nu, and Yunling Mountains), the Southeast Yunnan Mountains (Ailao and Wuliang Mountains), and the lower reaches of the Lancang River Basin (Simao District and Jinggu County). These areas maintain high ecosystem service values due to their unique terrain and climatic conditions. Cold spots, accounting for 25% of the total area, are mainly found in densely populated regions such as the Central Yunnan urban agglomeration, Zhaotong City in Northeast Yunnan, and the middle and upper reaches of the Jinsha River Basin (Heqing and Binchuan Counties). Non-significant areas account for 45% of the province and are scattered throughout. This spatial pattern of hotspots and cold spots provides an important foundation for the identification of ecological sources.

3.4. Identification of ESP

3.4.1. Identification of Ecological Sources

Using MSPA and landscape connectivity analysis, we identified ecological sources in Yunnan Province (Figure 9). In 2020, 56 ecological source patches (54,173.35 km²) were delineated, including 25 primary ecological sources and 31 secondary ecological sources. By 2030, the SSP1-1.9 scenario contained the largest ecological source area (55,825.57 km²) with 61 total patches, including 37 secondary sources—the highest number among all scenarios. The SSP2-4.5 scenario showed 57 patches (50,272.92 km²) with 32 secondary sources, while the SSP5-8.5 scenario also contained 57 patches (49,976.12 km²) but only 31 secondary sources.

In 2020, the spatial distribution of ecological sources under the three scenarios exhibited a clear pattern of “more in the west and less in the east.” First-class ecological sources were primarily located in Northwest Yunnan, the Nujiang River Basin, and the middle and lower reaches of the Lancang River Basin. Secondary sources varied across scenarios. In 2020, they were mainly distributed in the lower reaches of the Yuan River Basin and Funing County in eastern Yunnan. Under the SSP1-1.9 scenario, in addition to the lower reaches of the Yuan River Basin and Funing County, new secondary sources appeared in the Central Yunnan urban agglomeration. In contrast, the SSP2-4.5 and SSP5-8.5 scenarios resulted in reductions in ecological source areas. This distribution highlights a fundamental spatial dichotomy in Yunnan, driven by the interplay between physiographic and socio-economic factors. The western and southern regions—characterized by the rugged Hengduan Mountains and the well-preserved tropical rainforests of Xishuangbanna—function as critical ecological cores, providing key services such as biodiversity refugia, water conservation, and carbon sequestration. In contrast, the central and eastern parts have experienced intensive urban and agricultural expansion, leading to significant fragmentation, and squeezing of ecological spaces. Thus, the observed pattern not only reflects topographic and climatic gradients but also underscores the tension between socio-economic development and ecological conservation.

3.4.2. Results of Resistance Surface Construction

According to the results of resistance surface construction (Figure 10), low and relatively low resistance areas accounted for more than 50% of the province, mainly distributed in the border areas of Southern Yunnan, Eastern Yunnan, and the Yuan River Basin. These regions have significant ecological advantages: high forest coverage, suitable climate, superior ecological environment, and high ecosystem service efficiency. The middle-resistance areas are primarily concentrated in northwest Yunnan and the northern part of the Central Yunnan Urban Agglomeration. These areas are dominated by grassland and forest and have relatively low population density. Although there are certain restrictions on the ecosystem service function, the degree of ecological damage is relatively light. In contrast, although the areas with high resistance and relatively high resistance account for the smallest proportion (mainly distributed in the urban agglomeration in Central Yunnan and the surrounding cities in western Yunnan), their ecological pressure is the most significant: high intensity human activities lead to urban expansion and serious habitat fragmentation, which not only weaken the regional ecological carrying capacity, but also pose a serious obstacle to the formation and maintenance of ecosystem services.

3.4.3. Identification of Ecological Corridors, Pinch Points, and Barrier Points

1.

Ecological Corridor Identification

The ecological corridor network in Yunnan Province exhibits significant spatial heterogeneity (Figure 11). In 2020, a total of 132 ecological corridors were identified, including 38 key corridors (1119.38 km) and 94 potential corridors (6698.84 km). The key corridors are concentrated in the high-altitude area in Northwest Yunnan and the rainforest area in Southwest Yunnan, where the ecological sources are dense, there is a wide variety of animals and plants, and the information transmission and connectivity between organisms are high. These areas are key ecological protection areas and are irreplaceable; The potential corridors are mostly distributed in the central and eastern regions of Yunnan, where the development intensity is high. The population in this region is relatively dense, the degree of urban development is strong, and the resistance to the ecological environment is large, which makes the number of potential ecological corridors large, the distance is long, the protection width of corridors is small, and the replaceability is high.

Under different scenarios in 2030, SSP1-1.9 scenario features the largest number of corridors (149), the best connectivity, key corridors (33), and potential corridors (116), which are mainly connected between patches with large source areas in Northwest and South Yunnan; The number of corridors in SSP2-4.5 and SSP5-8.5 scenarios is similar (135 and 134), and the spatial distribution is similar. The corridors in western Yunnan are dominant, but the total length of corridors in the SSP5-8.5 scenario is the shortest due to the pressure of economic development. On the whole, the quality of the ecological corridor network presents the gradient characteristics of SSP1-1.9 > SSP2-4.5 > SSP5-8.5, reflecting the differential impact of different development scenarios on ecological connectivity.

2.

Ecological Pinch Points Identification

According to the identification results of ecological pinch points (Figure 12), we found: in 2020, 74 pinch points (99.026 km²) were identified, primarily located in northwestern and southern Yunnan, within well-functioning forest and grassland ecosystems. By 2030, the SSP1-1.9 scenario shows the highest number of pinch points (84; 107.607 km²), indicating the most robust ecological connectivity. In contrast, the SSP2-4.5 and SSP5-8.5 scenarios show reductions, with 55 and 60 pinch points, respectively. This gradient (SSP1-1.9 > SSP5-8.5 > SSP2-4.5) highlights how development pathways differentially affect ecological stability, underscoring the need to strengthen protection measures in these critical areas to maintain ecological function.

The detail results for 2030 scenarios are as follows: the SSP1-1.9 scenario contained the most pinch points (84; 107,607 km²), primarily in northwestern and southern Yunnan, indicating the best ecosystem stability. The SSP2-4.5 scenario contained 55 ecological pinch points (81,322 km²), primarily located in northwestern and southern border regions, mostly within forest and grassland areas. This represents the lowest number among all future scenarios, indicating a continued historical development trend that exacerbates environmental degradation and weakens ecological stability. In contrast, the SSP5-8.5 scenario registered 60 ecological pinch points (81,322 km²), with a spatial distribution similar to SSP2-4.5. Although the total area also decreased compared to 2020, the number ranks in the middle of the three scenarios. This intermediate outcome suggests that technical protection measures may partially offset the ecological pressures associated with intensive economic development.

3.

Ecological Barrier Points Identification

According to the identification results (Figure 13), the area of ecological barriers in Yunnan in 2020 is 48,448.98 km², mainly with moderate barriers (44%), mainly distributed in the urban agglomeration in Central Yunnan; High barrier areas are concentrated in Dali, Lijiang, Chuxiong, and Zhaotong. Under different climate scenarios in 2030, the barrier area of SSP1-1.9 scenario is the largest (52,042.950 km²), and the high barrier areas is transferred to the Jinsha River and the Pearl River Basin; In contrast, the total barrier areas under the SSP2-4.5 and SSP5-8.5 scenarios are smaller and are primarily distributed in Northwest Yunnan, the Central Yunnan Urban Agglomeration, and Lincang City, respectively. The priority for ecological restoration depends on the degree of barriers: high-barrier areas are difficult to improve due to human activities and climate impacts; moderate-barrier areas face significant pressure from urbanization; and low-barrier areas, mainly in western and eastern Yunnan, have the greatest potential for transformation. These low-barrier areas are considered the key targets for ecological improvement in this study.

In 2020, the area of the ecological improvement area was 17,855.100 km². By 2030, these improvement areas are projected to vary significantly under different climate scenarios: SSP1-1.9 (18,674.730 km²), SSP2-4.5 (15,658.650 km²), and SSP5-8.5 (14,359.500 km²). The improvement area is dominated by forest (>50%) and grassland, followed by cultivated land and construction land, and the proportion of water and unused land is the smallest. Because Yunnan Province is dominated by forest and grassland with high ecological value, its protection and transformation are more convenient. Although cultivated and construction lands face greater anthropogenic pressure, they remain amenable to improvement through targeted measures. In contrast, the limited and often fragmented water, and unused lands present greater restoration challenges. Therefore, the ecological restoration needs to adjust measures to local conditions and formulate a reasonable plan in combination with the characteristics of regional development.

3.4.4. Construction of ESP

Using the MSPA model and circuit theory, this study integrates identified ecological sources, corridors, pinch points, barrier areas, and improvement areas with the distribution of national protected areas in Yunnan Province to construct ecological security patterns (ESPs) under three climate scenarios for 2020 and 2030 (Figure 14). The results generally align with the spatial distribution outlined in Yunnan’s territorial spatial planning (2021–2035). The ESP under the SSP1-1.9 scenario closely resembles that of 2020, whereas the ESPs under SSP2-4.5 and SSP5-8.5 display comparable spatial distributions. Among the three scenarios, SSP1-1.9 exhibits the most favorable spatial configuration, reflecting a strong emission-reduction pathway dominated by green energy and resulting in enhanced ecological conditions. The SSP2-4.5 scenario, which most closely reflects the historical development trajectory, combines moderate economic growth with medium environmental governance capacity. This leads to noticeable differences from the 2020 ESP, particularly in the Central Yunnan urban agglomeration and the surrounding areas of cities in Northeast Yunnan. In contrast, the SSP5-8.5 scenario represents rapid economic growth accompanied by high greenhouse gas emissions, which exerts severe pressure on the ecological environment. Under this scenario, the overall effectiveness of the ESP is poor, with negative impacts extending beyond Central and Northeast Yunnan to also include the surrounding areas of cities in Western Yunnan.

4. Discussion

4.1. Optimization of ESP in Yunnan Province

By optimizing the results of ESP construction of three future climate scenarios in Yunnan Province in 2020 and 2030, it is found that the ESP of Yunnan Province is centered on forest-based ecological source, forming a corridor network relying on water, mountains, and forest belts, and presents a spatial structure of “three screens, two zones, six corridors and multiple points” (Figure 15).

The term “three screens” denotes the ecological barriers located along the southeastern edge of the Qinghai-Tibet Plateau in northwest Yunnan, the Ailao Mountain—Wuliang Mountain Ecological Barrier in Southwest Yunnan, and the southern border ecological barrier. Among them, Gaoligong Mountain, Ailao Mountain, and other mountains are natural ecological barriers. The mountain barrier serves as a framework to safeguard the plateau and forest ecosystems, enhance the protection of grasslands, wetlands, and wildlife, and promote the conservation and restoration of river basin ecological environments. Among them, the ecological barriers on the southeast edge of the Qinghai Tibet Plateau in Northwest Yunnan are mainly distributed in Diqing Tibetan Autonomous Prefecture, Nujiang Lisu Autonomous Prefecture, and the northwest urban areas of Lijiang, Dali, and Baoshan; The ecological barrier of Ailao Mountain Wuliang Mountain in Southwest Yunnan is mainly distributed in the middle and upper reaches of Yuan River- Hongshui River Basin; The southern border ecological barrier is concentrated in the border cities of southern Yunnan. The ecological sources identified in this paper are connected and concentrated in the Northwest and South Yunnan. The connectivity between the sources is good, and they become the ecological barriers to protect the ecological security and biodiversity of Yunnan Province.

The term “two zones” refers to the Yuanmou dry-hot valley region, dominated by the Jinsha River, and the karst region located in eastern and southeastern Yunnan (Nanpan River Basin). The two zones are relatively concentrated in spatial distribution. The basin and valley zone has developed into water networks and economic and cultural corridors. The river zone is used as the vein to strengthen the comprehensive control of rocky desertification, strictly protect the vegetation of the Rocky Mountains, afforestation, and grass planting, increase vegetation coverage, promote the quality of vegetation, and enhance the natural resilience of the regional ecosystem. Due to the rapid development of the urban agglomerations in Central Yunnan and cities in Northeast Yunnan, the ecological damage is significant, which makes the identified ecological sources fewer and the ecological corridors shorter. However, karst landforms such as stone forests and karst caves in the Jinsha River Basin and eastern Yunnan are important ecological resources that cannot be ignored. The ecological sources and corridors identified in eastern Yunnan in this study were systematically organized, resulting in the delineation of an ecological belt that safeguards Yunnan Province’s ecological resources.

“Six corridors” refer to the Nujiang River Biodiversity Conservation Corridor, Diqing-Lijiang Biodiversity Conservation Corridor, Yuan River Biodiversity Conservation Corridor, Pu’er-Xishuangbanna Biodiversity Conservation Corridor, Southeast Yunnan Soil and Water Conservation Corridor and Northeast-Central Yunnan Soil and Water Conservation Corridor, with the corridor as the network to ensure the smooth passage of species migration and the connectivity and continuity between important ecological function areas. Pu’er-Xishuangbanna Biodiversity Conservation Corridor, Yuan River Biodiversity Conservation Corridor and Southeast Yunnan Soil and Water Conservation Corridor play an important role in the protection of biodiversity in Southern Yunnan; Nujiang Biodiversity Conservation Corridor and Diqing-Lijiang Biodiversity Conservation Corridor jointly protect the ecological security of Northwest Yunnan; Due to the less identified ecological sources and corridors in eastern Yunnan under the SSP2-4.5 and SSP5-8.5 scenarios, combined with the identified ecological sources and corridors in 2020 and SSP1-1.9 scenarios, the Southeast Yunnan Soil and Water Conservation Corridor and the Northeast Central Yunnan Soil and Water Conservation Corridor are obtained. Although the connectivity is poor and the corridor length is short, they are of great significance for the ecological protection of Northeast Yunnan.

“Multiple points” refers to the comprehensive selection of ecological pinch points and improvement areas identified in 2020 and 2030 as ecological node protection areas with the main task of improving the ecological environment, which are mainly distributed in the northwest and south of Yunnan. Due to the rapid development of cities of the urban agglomeration in Central Yunnan and Western Yunnan, the ecological pinch points are relatively scattered, with multipoint as the core, polarized.

4.2. Ecological Security and Protection Strategy

4.2.1. Ecological Source Protection Strategy

The ecological source areas identified in this study represent the hotspots of ecosystem service value (ESV) in Yunnan Province. Dominated by forest and grassland land-use types, these areas form the core barriers for regional ecological security protection. To maintain the integrity and authenticity of these areas, the following protection strategies are proposed:

Firstly, we should strictly control the interference of human activities, implement differentiated control for different land types, giving priority to afforestation projects in mountainous areas, implement grazing prohibition policies on grasslands, build ecological buffer zones around water bodies, maintain appropriate use of cultivated land, and implement intensive development of construction land.

Second, by undertaking ecological rehabilitation efforts, including reforesting farmland and restoring grasslands and wetlands, vegetation cover is expected to recover progressively, thereby enhancing critical ecosystem services such as water retention and soil preservation.

Thirdly, establish and improve the monitoring and early warning system of ecological source disasters, focusing on the prevention and control of ecological risks such as geological disasters and soil erosion. Fourth, strictly restrict the change of the nature of the ecological source land and prevent the disorderly expansion of urban construction land from damaging the integrity of the ecological landscape.

In particular, it should be emphasized that key protection and restoration projects should be implemented for the source areas with important ecological functions, such as mountains and valleys in Northwest Yunnan, tropical rainforests in Southwest Yunnan, and mountains in Northeast Yunnan. For the nature reserves in the ecological source areas, the intensity of tourism development should be strictly controlled to avoid the transitional construction of tourism facilities; Improve the protection management system and formulate a scientific protection plan; We will strengthen the protection of biodiversity and focus on protecting endangered species and their habitats. By building a comprehensive protection network, we can effectively maintain the regional ecological security pattern and realize the sustainable supply of ecosystem services.

4.2.2. Ecological Corridor Protection Strategy

According to the identification results of ecological corridors, we found that ecological corridors in Yunnan are frequently degraded by adjacent human activities, reducing connectivity and hindering species migration. To counter this, we recommend enhancing corridor construction and restoration to improve ecological function and minimize disturbance. Key actions include integrating corridor protection into urban planning with a “zero-occupancy” policy for critical segments, utilizing ecological bridges, and optimizing vegetation in pinch points. Given regional ecosystem heterogeneity, tailored strategies are essential: designing corridors based on local land use, topography, and climate; protecting key river and forest corridors from agricultural and urban expansion; implementing vegetation restoration; and establishing long-term monitoring. Prioritizing the preservation of original vegetation will maintain natural ecological characteristics, thereby improving network connectivity and stability to support regional biodiversity conservation.

4.2.3. Protection Strategies for Ecological Pinch, Barrier Area, and Improvement Area

The ecological pinch in Yunnan Province exhibits a decentralized distribution pattern characterized by “higher concentrations in the north and south and lower concentrations in the east and west,” primarily occurring in areas with relatively intact ecological functions, such as forests and grasslands. These pinch points are very important to maintain environmental stability and ecological security, but some pinch points are located in the improvement area, indicating that they face high risks and need to be protected. It is suggested to reduce human interference around key pinch points such as nature reserves, wetland parks, and forest parks, repair degraded areas, and improve ecological functions.

Ecological barrier areas are primarily located in regions experiencing rapid urbanization and high population density. Intense human activities in these regions have caused substantial disturbances to natural ecosystems, severely impeding the normal migration and dispersal of species. For infrastructure barriers such as roads and buildings that are difficult to remove, small ecological stepping stones can be set in key ecological areas, and underground passages can be constructed as a biological corridor, and the spatial layout of urban green space can be optimized to build a microhabitat network. In ecological areas located between mountains, rivers, or lakes, additional green space can be added between adjacent ecological patches. Vegetation restoration project should be implemented, with a focus on improving the landscape connectivity—particularly in the northeastern region; build a multi-level ecological buffer zone.

Ecological improvement areas are primarily located within forested regions and require targeted transformations. Key measures include implementing the project of returning sloping farmland to forest; strictly controlling agricultural expansion in soil erosion risk zones; promoting afforestation on suitable barren mountains; and enhancing the ecological functions of farmland in protected areas. For other ecologically vulnerable areas, differentiated strategies should be adopted, such as enforcing a grass-livestock balance system and restoring grassland vegetation through grazing restrictions. Around construction land, optimizing the green space system by “inserting green space at the seams” and promoting multi-layered, native-vegetation-based greening can strengthen ecological connectivity and enhance overall ecosystem services.

4.3. Limitations of the Study

Human activities and land use changes can significantly alter landscape patterns, resulting in decreased ecosystem service values and increased ecological vulnerability. The scientific construction of the ecological security pattern can effectively optimize the ecological supervision system, reasonably allocate natural resources, and improve the green infrastructure network, so as to comprehensively improve the ecosystem service function. This study, based on the theoretical framework of the “source–resistance–corridor” approach and integrating ecosystem service value assessment, MSPA, and circuit theory, uses Yunnan Province as a case study to propose a novel method for constructing ecological security patterns. It not only offers a new perspective for regional ecological protection but also provides a scientific basis for land spatial planning and decision-making. However, there are still some limitations. (1) This paper uses the equivalent factor table for ecological value developed by Xie Gaodi as the basis for evaluating the value of ecological services. Still, the table is based on the national average level. Although it is modified according to the distribution of land types in the study area, it cannot fully reflect the actual characteristics of the ecological environment in the study area. Therefore, future research should further refine and calibrate the ecological value equivalent factors based on the local environmental context to enhance the accuracy and applicability of the evaluation results. This refinement will enable a more precise reflection of the distinctive characteristics of regional ecosystem services and offer a more robust theoretical foundation for ecological conservation and spatial planning. (2) In the study of constructing the ecological security pattern in Yunnan Province, this paper mainly extracts the ecological corridor through the ecological resistance value, but this method has not fully considered the impact of urban economy, human activities and other factors on the ecological corridor, In future studies, we will incorporate more representative human activity data, such as nighttime light and industrial infrastructure distribution, to improve the construction and optimization of resistance surfaces and the selection and setting of the ecological corridor width is not accurate enough. The constructed ecological security pattern is relatively macro, and future research can pay more attention to the micro level to realize the optimization of the ecological security pattern at multiple levels and scales. (3) Furthermore, the static ecosystem service value assessment method used in this study is difficult to capture the potential nonlinear value decay caused by dynamic feedback such as biodiversity loss; Secondly, the identification of ecological improvement areas did not fully consider socio-economic constraints, which may overestimate the actual restoration capacity. Future research needs to develop a dynamic model of ecosystem service value and integrate socio-economic feasibility analysis to enhance the practical effectiveness of ecological restoration planning.

5. Conclusions

This study establishes an innovative framework that integrates multi-scenario land use simulation, ecosystem service value (ESV) assessment, and the construction and optimization of ecological security patterns (ESP). The framework not only enables dynamic simulation of future land use and ecosystem services in Yunnan Province under climate change scenarios, but also identifies and delineates key ecological sources, corridors, pinch points, and barrier areas, thereby proposing differentiated strategies for territorial ecological security protection. Based on this framework, the main conclusions are as follows:

(1) From 2000 to 2030, the overall ranking of ecosystem service values (ESV) in Yunnan Province follows the order: regulating services > supporting services > provisioning services > cultural services. During 2000–2020, the ESV showed a decline followed by an increase, with the greatest growth observed in water bodies (253.68 million). Under all three future climate scenarios, the ESV is projected to increase compared with 2020, with the largest increase under SSP2-4.5, followed by SSP5-8.5 and SSP1-1.9, primarily driven by rising values of water bodies and grasslands. Spatially, the distribution of ESV remains consistent, showing a pattern of “high in the central region, low in the east and west.” High-value areas are concentrated in northwestern and southwestern Yunnan, while low-value areas are distributed around the central and western urban clusters and continue to expand outward. The ESV shows positive spatial correlations, with Moran’s I of first-level services ranked as: regulating services > supporting services > cultural services > provisioning services. Hotspot areas are mainly distributed in the southern Yunnan plateau mountains and river basins, while coldspot areas are concentrated in central, northeastern, and western Yunnan.

(2) In 2020, 56 ecological sources were identified in Yunnan Province; under SSP1-1.9, 61 sources; under SSP2-4.5, 57 sources; and under SSP5-8.5, 57 sources. These are mainly distributed in northwestern Yunnan, the Nujiang River Basin, and the Lancang River Basin, forming a pattern of “more in the west, fewer in the east,” with forests and grasslands as the dominant land cover types.

(3) In 2020, 132 ecological corridors and 74 pinch points were identified. Under SSP1-1.9, there are 149 corridors and 84 pinch points; under SSP2-4.5, 135 corridors and 55 pinch points; under SSP5-8.5, 134 corridors and 60 pinch points. Key ecological corridors are primarily located in northwestern and southwestern Yunnan, while potential corridors are concentrated in central Yunnan. The pinch points exhibit a distribution characterized by “more in the north–south and fewer in the east–west,” and are mainly concentrated in northwestern and southern Yunnan. The ecological barrier area in 2020 was 48,448.98 km², with an improvement area of 17,855.1 km². Under SSP1-1.9, the barrier area expands to 52,042.95 km² with an improvement area of 18,674.73 km²; under SSP2-4.5, 39,247.38 km² and 15,658.65 km², respectively; and under SSP5-8.5, 36,670.23 km² and 14,359.5 km², respectively. Barrier areas are mainly distributed in densely populated and rapidly urbanizing cities, while improvement areas are found in northwestern, southwestern, and eastern Yunnan, where environmental restoration capacity is stronger.

(4) Among the scenarios, SSP1-1.9 (strong emission reduction) yields the most favorable ecological pattern, while SSP2-4.5 (medium emissions) achieves a balance between economic development and environmental management, and SSP5-8.5 (high emissions) produces the least favorable pattern. Based on multi-scenario simulations, a refined ecological security framework characterized by “three shields, two belts, six corridors, and multiple nodes” was developed. Corresponding conservation strategies were designed for ecological sources, corridors, pinch points, barrier areas, and restoration zones to enhance ecological security in Yunnan Province.